Though Language Models Err While They Strive: Conformal Prediction for Self-Correcting Scientific Generation

• URL: http://arxiv.org/abs/2607.16704v1
• Date: Sat, 18 Jul 2026 08:38:19 GMT
• Title: Though Language Models Err While They Strive: Conformal Prediction for Self-Correcting Scientific Generation
• Abstract summary: We introduce Scientific Feasibility Control SFC, a graph-structured conformal prediction framework.<n>SFC decomposes scientific reasoning into atomic absolute-coherent-factuality units.<n>We demonstrate SFC across established scientific reasoning benchmarks including PhyX multimodal physics, MATH, ScienceQA, and ARC Challenge.
• Score: 4.019632358747631
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models frequently violate fundamental scientific principles when generating technical content, undermining their reliability in scientific applications. We introduce Scientific Feasibility Control SFC, a graph-structured conformal prediction framework that provides statistical guarantees for scientific reasoning validity through progressive absolute-coherent-factuality validation. Our approach decomposes scientific reasoning into atomic absolute-coherent-factuality units requiring both individual correctness against physical laws and logical substantiation from preceding context, addressing the cascade effect where early scientific errors contaminate subsequent reasoning steps. Unlike independence-based methods that treat claims in isolation, SFC models logical dependencies as approximate deducibility graphs and operates through real-time validation with dynamic branching when scientific violations are detected, the system branches to alternative generation paths using verified context as foundation. We demonstrate SFC across established scientific reasoning benchmarks including PhyX multimodal physics, MATH, ScienceQA, and ARC Challenge, achieving 50.1 percent accuracy on PhyX physics reasoning, substantially outperforming recent reasoning models including DeepSeek-R1 49.8 percent and GPT-4 45.8 percent while providing 91.7 percent scientific validity with formal conformal coverage guarantees at alpha equals 0.10 confidence level and reducing scientific law violations by 73 percent across multiple model architectures.

Related papers

• Accurate, Interdisciplinary and Transparent Structure-property Understanding with Deep Native Structural Reasoning [103.788281428856]
  We introduce SciReasoner, a scientific foundation model for native structural reasoning across proteins, small molecules and inorganic crystals.<n>SciReasoner discretizes coordinates, topologies and periodic connectivities into a unified structure-aware vocabulary.<n>Across 86 benchmarks, SciReasoner achieves state-of-the-art performance on 67 tasks.
  arXiv  Detail & Related papers  (2026-07-08T17:59:59Z)
• Knowing When Not to Answer: Abstention-Aware Scientific Reasoning [2.680633756465714]
  In scientific settings, unsupported or uncertain conclusions can be more harmful than abstaining.<n>We study this problem through an abstention-aware verification framework.<n>We evaluate this framework across two scientific benchmarks: SciFact and PubMedQA.
  arXiv  Detail & Related papers  (2026-02-15T15:29:43Z)
• SciIF: Benchmarking Scientific Instruction Following Towards Rigorous Scientific Intelligence [60.202862987441684]
  We introduce scientific instruction following: the capability to solve problems while strictly adhering to the constraints that establish scientific validity.<n>Specifically, we introduce SciIF, a multi-discipline benchmark that evaluates this capability by pairing university-level problems with a fixed catalog of constraints.<n>By measuring both solution correctness and multi-constraint adherence, SciIF enables finegrained diagnosis of compositional reasoning failures.
  arXiv  Detail & Related papers  (2026-01-08T09:45:58Z)
• PRISM-Physics: Causal DAG-Based Process Evaluation for Physics Reasoning [57.868248683256574]
  PRISM-Physics is a process-level evaluation framework and benchmark for complex physics reasoning problems.<n> Solutions are represented as directed acyclic graphs (DAGs) of formulas.<n>Results show that our evaluation framework is aligned with human experts' scoring.
  arXiv  Detail & Related papers  (2025-10-03T17:09:03Z)
• SCI-Verifier: Scientific Verifier with Thinking [37.08904000514563]
  Large language models (LLMs) are increasingly applied to scientific reasoning.<n>Existing verification studies in scientific domains suffer from two major limitations.<n>We propose solutions at both the data and model levels.
  arXiv  Detail & Related papers  (2025-09-29T04:58:43Z)
• SciReasoner: Laying the Scientific Reasoning Ground Across Disciplines [112.78540935201558]
  We present a scientific reasoning foundation model that aligns natural language with heterogeneous scientific representations.<n>The model is pretrained on a 206B-token corpus spanning scientific text, pure sequences, and sequence-text pairs, then aligned via SFT on 40M instructions.<n>It supports four capability families, covering up to 103 tasks across: (i) faithful translation between text and scientific formats, (ii) text/knowledge extraction, (iii) property prediction, (iv) property classification, (v) unconditional and conditional sequence generation and design.
  arXiv  Detail & Related papers  (2025-09-25T17:52:06Z)
• Bayesian Epistemology with Weighted Authority: A Formal Architecture for Truth-Promoting Autonomous Scientific Reasoning [0.0]
  This paper introduces Bayesian Epistemology with Weighted Authority (BEWA)<n>BEWA operationalises belief as a dynamic, probabilistically coherent function over structured scientific claims.<n>It supports graph-based claim propagation, authorial credibility modelling, cryptographic anchoring, and zero-knowledge audit verification.
  arXiv  Detail & Related papers  (2025-06-19T04:22:35Z)
• Atomic Reasoning for Scientific Table Claim Verification [83.14588611859826]
  Non-experts are susceptible to misleading claims based on scientific tables due to their high information density and perceived credibility.<n>Existing table claim verification models, including state-of-the-art large language models (LLMs), often struggle with precise fine-grained reasoning.<n>Inspired by Cognitive Load Theory, we propose that enhancing a model's ability to interpret table-based claims involves reducing cognitive load.
  arXiv  Detail & Related papers  (2025-06-08T02:46:22Z)
• Generating Scientific Claims for Zero-Shot Scientific Fact Checking [54.62086027306609]
  Automated scientific fact checking is difficult due to the complexity of scientific language and a lack of significant amounts of training data. We propose scientific claim generation, the task of generating one or more atomic and verifiable claims from scientific sentences. We also demonstrate its usefulness in zero-shot fact checking for biomedical claims.
  arXiv  Detail & Related papers  (2022-03-24T11:29:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.